Revisiting Weakly Supervised Pre-Training of Visual Perception Models

作者：Mannat Singh, Laura Gustafson, Aaron Adcock, Vinicius de Freitas Reis, Buğra Gedik, Raj Prateek Kosaraju, Dhruv Mahajan, Ross Girshick, Piotr Dollár, Laurens van der Maaten · 发表于：2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · 年份：2022 · DOI：10.1109/cvpr52688.2022.00088 · 被引用次数：91 · 研究领域：Domain Adaptation and Few-Shot Learning、Multimodal Machine Learning Applications、Advanced Image and Video Retrieval Techniques

Model pre-training is a cornerstone of modern visual recognition systems. Although fully supervised pre-training on datasets like ImageNet is still the de-facto standard, recent studies suggest that large-scale weakly supervised pretraining can outperform fully supervised approaches. This paper revisits weakly-supervised pre-training of models using hashtag supervision with modern versions of residual networks and the largest-ever dataset of images and corresponding hashtags. We study the performance of the resulting models in various transfer-learning settings including zero-shot transfer. We also compare our models with those obtained via large-scale self-supervised learning. We find our weakly-supervised models to be very competitive across all settings, and find they substantially outperform their self-supervised counterparts. We also include an investigation into whether our models learned potentially troubling associations or stereotypes. Overall, our results provide a compelling argument for the use of weakly supervised learning in the development of visual recognition systems. Our models, Supervised Weakly through hashtAGs (SWAG), are available publicly.